In recent years there has been considerable commercial interest in the vapor phase process disclosed in U.S. Pat. No. 3,190,912 for the preparation of unsaturated organic esters such as vinyl acetate from olefinic compounds, organic acids, and oxygen. Palladium metal-containing catalysts supported on an alumina carrier have been found to be particularly useful in such a process for the preparation of vinyl acetate from a gaseous mixture of ethylene, acetic acid, and oxygen. The activity and/or stability of such catalyst compositions have been enhanced by the use of various activators including alkali metal acetates, especially sodium and potassium acetate as well as certain metals such as gold and platinum. U.S. Pat. No. 3,190,912 also discloses the manufacture of other unsaturated esters by varying the olefinic compound and carboxylic acid reactants. The preparation of allyl acetate by related processes is further described in U.S. patent specification Ser. No. 330,536 which was published in the Official Gazette, page 1640, Jan. 28, 1975, under the trial Voluntary Protest Program.
Despite the improvements in activity and/or stability of the catalyst composition that have been attained heretofore by the use of various activators, the Group VIII noble metal catalyst compositions employed in the vapor phase processes still undergo a gradual loss in activity under prolonged use and consequently require periodic regeneration in order to maintain the desired activity. Methods designed to overcome the loss in catalytic activity are set forth in U.S. Pat. Nos. 3,650,983 and 3,879,311 which are directed to a regeneration procedure for a Group VIII noble metal catalyst, supported on an alumina carrier, used in vapor phase vinyl acetate process.
It has now been found that catalyst activity is not the sole criterion for determining catalyst life, and that the prior proposals to reactivate the catalyst composition by washing techniques or by regeneration will not assure long catalyst life. More specifically, it was found that during use an alumina loss from the catalyst composition takes place at a constant rate and that the alumina carrier exhibits a serious loss of crush strength. Not only is there a marked decrease in the average crush strength, but the percentage of alumina having less than the prescribed minimum crush strength increased dramatically under prolonged use, for example, in the vapor phase synthesis of vinyl acetate. In fact, the loss of crush strength is so significant that the useful life of the Group VIII noble metal catalyst can actually be limited by the decreasing crush strength. It was further determined that the loss of alumina from the carrier and the attendant drop in the physical strength of the carrier were caused by the presence of a lower carboxylic acid in the feed mixture, e.g., acetic acid, in the vinyl acetate vapor phase processes. This acid reacts with the alumina under process conditions to form aluminum salts and in the periodic reactivation, as set forth in U.S. Pat. No. 3,650,983, these salts are removed with the wash liquor. Attempts to overcome this problem by techniques which might stabilize the presently employed alumina carriers against acetic acid attack were unsuccessful. Moreover, efforts to find a more stable carrier were hampered by the lack of knowledge as to those factors which determine whether the catalyst carrier will have the desired stability when in use.
An improved alumina carrier for a catalyst composition employed in the vapor phase synthesis of vinyl acetate is the subject of U.S. Pat. No. 3,567,767. This patent prescribes the use of a high purity alumina, the purity being no lower than 99%, having a surface area of 60 to 150 m.sup.2 /g. The palladium metal catalyst supported on such an alumina is stated to have an improved and high catalytic activity. Although improved catalyst life ("stability" as defined for the present invention) is mentioned, this advantage is related to the use of low partial pressures of the organic carboxylic acid and oxygen during the synthesis rather than to the improved alumina carrier. The use of low acid partial pressures for synthesis is not recommended, since it not only decreases the production rate but can put the reaction feed gas mixture in the explosive range.
One use of high purity alumina carriers in catalyst compositions for the vinyl acetate vapor phase process are also disclosed in Japanese Patent Publication Nos. 24873, 24874 and 24876 all of which were published on July 7, 1972. These patent publications cover a variety of the so-called "high purity" aluminas and catalysts prepared on these carriers are stated to have improved catalytic activity and/or stability. Thus, for example, Japanese Patent Publication No. 24,873 is concerned with an alumina containing substantially no delta-alumina and of 95% or greater purity. Although both the theta and alpha forms of alumina can be present, the use of an alumina carrier where all the alumina is in alpha form is also described. Furthermore, this patent publication calls for an alumina carrier having a surface area of from 5 to 120 m.sup.2 /g and a pore volume of 0.20 to 0.40 cc/g. Japanese Patent Publication No. 24,874 achieves improved catalyst activity in the vinyl acetate synthesis by utilizing as the carrier silica-alumina having an alumina content of from 50 to 95%, a surface area of 10 to 150 m.sup.2 /g, a pore volume of 0.3 to 0.7 cc/g, and an alpha-alumina content of less than 5%. On the other hand, Japanese Patent Publication No. 24,876 calls for the use of an alumina carrier of 95% or greater purity having a delta-alumina content of 5% or less, a surface area of from 5 to 120 m.sup.2 /g, a pore volume of 0.20 to 0.40 cc/g, a particle diameter of from 2 to 6 mm, and wherein 90% or more of the palladium metal is located in the area which is up to 10% from the surface of the carrier. Carrier stability under actual process conditions is not mentioned, or implied, in any of the foregoing patents. As previously stated, catalysts of the type described when used in the synthesis process require periodic reactivation or regeneration. During synthesis and reactivation or regeneration these catalysts are exposed to conditions which encourage carrier degradation. Experience has shown that carriers of high purity alumina with a low surface area and a high crystalline alpha-alumina content are most resistent to degradation. For example, carriers containing silica cannot be reactivated according to U.S. Pat. No. 3,650,983 without considerable loss of silica, and carriers containing more than 2 to 5% theta-alumina undergo rapid attack by carboxylic acids which causes swelling of the carrier and plugging of the reactor. Even alpha-alumina undergoes attack by carboxylic acids, but the rate is considerably slower if the alumina crystallites are large (high degree of crystalline alpha-alumina).
Another patent pertaining to a palladium metal-containing catalyst supported on an alumina carrier is U.S. Pat. No. 3,883,442. Here the modified alumina support is prepared by impregnating a high surface area alumina with a boron compound which upon calcination yields B.sub.2 O.sub.3, drying the impregnated alumina, and then calcining it at a temperature within the range of 675.degree. to 1400.degree. C. The alumina carrier thus prepared has a surface area between about 5 to 100, preferably about 20 to 70 m.sup.2 /g, and a porosity of about 0.2 to 0.8, preferably 0.3 to 0.7 ml/g. In the preceding patent the addition of the boron compounds is directed toward preventing the conversion of the alumina to the highly crystalline alpha form while maintaining a high surface area. Such an alumina would not meet the carrier requirements set forth in the present invention.
Each of the aforementioned patents emphasize the importance of the method of manufacturing the alumina carrier. The wide variety of proposals in these patents also suggest that the basis for the selection of those alumina carrier physical and chemical properties which would ensure the desired improvement in catalyst life is not readily apparent and furthermore that the selection of the right combination of properties is extremely difficult.